lkml.org 
[lkml]   [2001]   [Jul]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: CPU affinity & IPI latency

On 13-Jul-2001 Larry McVoy wrote:
> Be careful tuning for LMbench (says the author :-)
>
> Especially this benchmark. It's certainly possible to get dramatically
> better
> SMP numbers by pinning all the lat_ctx processes to a single CPU, because
> the benchmark is single threaded. In other words, if we have 5 processes,
> call them A, B, C, D, and E, then the benchmark is passing a token from
> A to B to C to D to E and around again.
>
> If the amount of data/instructions needed by all 5 processes fits in the
> cache and you pin all the processes to the same CPU you'll get much
> better performance than simply letting them float.
>
> But making the system do that naively is a bad idea.

Agree.


>
> This is a really hard area to get right but you can take a page from all
> the failed process migration efforts. In general, moving stuff is a bad
> idea, it's much better to leave it where it is. Everything scales better
> if there is a process queue per CPU and the default is that you leave the
> processes on the queue on which they last run. However, if the load average
> for a queue starts going up and there is another queue with a substantially
> lower load average, then and ONLY then, should you move the process.

I personally think that a standard scheduler/cpu is the way to go for SMP.
I saw the one IBM guys did and I think that the wrong catch there is trying
always to grab the best task to run over all CPUs.
I think that this concept could be relaxed introducing less chains between each
CPU scheduler.
A cheap load balancer should run, "time to time"(tm), to move tasks when a
certain level of unbalancing has been reached.
This will give each scheduler more independence and will make it more scalable,
IMVHO.


> This is an area in which I've done a pile of work and I'd be interested
> in keeping a finger in any efforts to fix up the scheduler.

We've, somehow, understood it :)



- Davide

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 12:57    [W:0.082 / U:0.148 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site